Intended for healthcare professionals

Rapid response to:

Clinical Review State of the Art Review

HPV testing as a screen for cervical cancer

BMJ 2015; 350 doi: (Published 30 June 2015) Cite this as: BMJ 2015;350:h2372

Rapid Response:

HPV testing vs Wilson's principles. Does it stand the test?

Annekathryn Goodman (June 30 2015, 2015;350:h2372) provides a detailed review of Human Papilloma Virus (HPV) testing presenting it as a screening test for cervical cancer. This is an important area of public health given that a blood test could potentially abate repetitive, intimate examinations for women across the world. Moreover, the HPV testing appears to meet Wilson’s criteria for a good screening program [1].
Annekathryn Goodman has listed the burden of HPV and its consequences. This meets Wilson’s first condition that the health problem should be important. The pathogenicity, treatment and the benefits of early management of HPV are well established, similarly encompassed in Wilson’s criteria. While not made explicit in this article, we infer that HPV testing could revolutionise the accessibility of women to cervical health.

However, can a woman rely simply on a HPV test to determine whether or not she is applicable for with routine potentially lifesaving smears?

The crux of the matter comes to the accuracy and performance of this test. Annekathryn Goodman relays this information by way of positive and negative predictive values, but such measures of performance are not solely dependent on the intrinsic values of the test. For example, the prevalence of the condition also has a bearing on how well the test appears to work [2].

Nonetheless, even if this was a brilliant test, it is only going to pick up what you want it to pick up. It is all well and ood if the HPV subtypes 16 and 18 (or 45) are the only strains involved in cervical cancer. Furthermore, I calculated A. Goodman to have grossly underestimated the expense of the test. The test quoted by Qiagen comes in the form of a 96 test kit which costs approx. £350, which equates to the £3.30 per test that A. Goodman has proposed. However, if you deliberate that for one HPV test, a sample will need to be repeated in triplicate for reliability, and then repeated again (at least in duplicate) for the various strains. Additionally there are necessary tests run alongside for quality control and assurance. Furthermore, nucleic material must be prepared from each blood sample, and that a technician will need to operate and maintain analysers associated with these tests. A cost of £50 per sample would be more realistic. Wilson’s criteria are seldom met by DNA screening tests and the price tag is the big elephant in the room underpinning the scarcity of such screening tests.

[1] Wilson JMG, Jungner G. Principles and practice of screening for disease. Geneva: WHO; 1968. accessed 05/07/2015

[2] 4. Measurement error and bias. The BMJ, 2015. accessed 05/07/2015

Competing interests: No competing interests

06 July 2015
Balraj Bishan Singh Mavi
Medical Student
Jeetesh V Patel, Sameera B Mavi
The University of Nottingham